A discrepancy between cytology and bacterial culture methods is sometimes observed in canine otitis externa. The objective of this study was to compare results from cytology, bacterial culture and 16S amplicon profiling. Twenty samples from 16 dogs with chronic suppurative otitis externa were collected. A direct cytological evaluation was carried out during the consultations. Aerobic bacterial culture and susceptibility were performed by an external laboratory used in routine practice. For 16S amplicon profiling, DNA was extracted and the hypervariable segment V1−V3 of the 16S rDNA was amplified and then sequenced with a MiSeq Illumina sequence carried out by the Mothur software using the SILVA database. A good correlation between cytology and bacterial culture was observed in 60% of the samples. Some bacterial species revealed by bacterial culture were present with low relative abundance (<10%) in 16S amplicon profiling. Some bacterial species revealed by the 16S amplicon profiling analysis were not identified with culture; most of the time, the offending species was a Corynebacterium. To conclude, a careful interpretation of the results of bacterial culture should be made and always be in agreement with the cytology. The 16S amplicon profiling method appears to be a more sensitive method for detecting strains present in suppurative otitis but does not provide information on bacterial susceptibility.

Otitis media can be a consequence of chronic otitis externa and could represent a perpetuating factor. While the microbiota of the EEC in healthy dogs and in the presence of otitis externa has been described, only sparse information is available concerning the normal microbiota of the middle ear. The objective was to compare the tympanic bulla (TB) with the external ear canal (EEC) microbiota in healthy dogs. Six healthy experimental Beagle dogs were selected based on the absence of otitis externa, negative cytology and bacterial culture from the TB. Samples from the EEC and TB were collected directly after death using a total ear canal ablation and lateral bulla osteotomy. The hypervariable segment V1-V3 of the 16S rDNA was amplified and sequenced with a MiSeq Illumina. The sequences were analyzed by the Mothur software using the SILVA database. No significant differences between the EEC and TB microbiota for the Chao1 richness index (p = 0.6544), the Simpson evenness index (p = 0.4328) and the reciprocal Simpson alpha diversity (p = 0.4313) were noted (Kruskal-Wallis test). A significant difference (p = 0.009) for the Chao1 richness index between the right and left EEC was observed. The microbiota profile was similar in the EEC and the TB of the Beagles.